advice on software analysis fem

[f_lambiase]

Hello everyone, I am a mechanical engineer and for the first time I find myself having to deal with the fem analysis and it is the first time I write on this forum.
work in a company that builds railway wagons that travel in interrupting regime (type carri plasser).
as for the creation of technical files is required/preferred that the frame is verified with fem procedure, we are evaluating which software to buy and of course if there is the possibility to do a course however based on our needs, then checks of welded frames according to the current regulations.
I would need someone to tell me more or less about a software quite immediate and not too sophisticated by the cost more or less contained, taking into account that the 3d are made outside by a designer who uses inventor, so I should import the files modeling them and realize the analyses.
thanks to everyone in advance.

 

[The_Matrix]

Hello everyone, I am a mechanical engineer and for the first time I find myself having to deal with the fem analysis and it is the first time I write on this forum.
work in a company that builds railway wagons that travel in interrupting regime (type carri plasser).
as for the creation of technical files is required/preferred that the frame is verified with fem procedure, we are evaluating which software to buy and of course if there is the possibility to do a course however based on our needs, then checks of welded frames according to the current regulations.
I would need someone to tell me more or less about a software quite immediate and not too sophisticated by the cost more or less contained, taking into account that the 3d are made outside by a designer who uses inventor, so I should import the files modeling them and realize the analyses.
thanks to everyone in advance.

there are many pre/post/solutors on the market.
some are "generic", but very well known, flexible and powerful.. .
others are "specific", therefore dedicated and rigid.. .

in our suite, as a solutor there is always nx/nastran (a guarantee), while as a pre / place you can choose from 2 different solutions:nx/advanced simulation: high-end solution, very integrated with the cad, so simple to use (you need to read the inv files in cad nx)femap: high-end solution, stand-alone

 

[f_lambiase]

thanks for the answer but I am not very experienced, when you talk about pre and post solutors you mean design 3d and analysis software? ...I called ansys italia and today I have an appointment, by phone they told me that in my case I would need ansys professional nls, what do you think? ....in the end I should simply carry out static checks in linear compo.. but I had read also of nastran, do you think it is best as a value for money?

 

[Matteo]

hi the section is not the most suitable (there is a section dedicated to fem/cfd) but I try to answer you.

If you are looking for a "simple" solution (though not to underestimate the fact that the fem analysis is for specialists, designer does not necessarily mean strutturist) and "rapida" the milgior thing is to evaluate an integrated fem in the cad. in this field the point of reference is undoubtedly promechanic, with the fem modules of solidworks and catia immediately behind (I do not speak of "power of the instrument", but of degree of integration). I don't know the nx proposal, but I think it's good.

the problem in your case is that you already have a cad: I read that you use inventor, so I would also evaluate algor (I think that now is called inventor simulation mechanical): it is not completely integrated (we are at the lviello of "switch") and there are some nemmno entities too advanced, but if they give it to you at a super price, considering that you are not "specialists" in the fem, that things too complex however it is better not to manage them internally, it could go.

as reported "the matrix" among unintegrated femap is a small jewel (it was one of the first pre/posts born for windows): not too bright for "enormous" models, but definitely very versatile. I know it associated with neinastran (not nxnastran) but for your needs I don't think it makes much difference.

for any information are here (for transparency: we are suppliers of calculation and simulation services).

 

[The_Matrix]

thanks for the answer but I am not very experienced, when you talk about pre and post solutors you mean design 3d and analysis software? . . .I called ansys italia and today I have an appointment, by phone they told me that in my case I would need ansys professional nls, what do you think? ....In the end I should simply carry out static audits in linear compo.. but I had read also of nastran, do you think it is better as a value for money?

I am a solution provider, so it is not correct that I express any judgment.
Hi.

 

[Matteo]

chapter ansys: definitely a good product, especially complete, but with the following weaknesses.

- price
- a little flourishing.
- not integrated with the cad (commercially it is proposed as integrable, but they are just switches. it goes a little better integration with spaceclaim, proposed bundled, possibly fouling to ansysy that often forgets, but sycnerously it is not true integration even that)
- expandable to worlds that I think you are not interested in (cfd)
- long training course

if you are interested exclusively in mechanics and if you choose a pre/post stand alone, then better something more "smart" as femap.

 

[f_lambiase]

First of all, thank you for your concern, and apologize for the section but I am not yet practical.
I have met ansys, practically as the analysis I have to carry out is in the linear and static field, it offers us:
- the basic package ansys designspace
-the module to create 3d or change them/adjust them as not always the designer (person outside the company) creates a good design to pass immediately to the analysis
- will follow me step by step in a project that I have to accomplish so as to create a personalized course

all to a figure much less than 20000 euros (I am not accurate because I do not know if it is correct to talk about prices on the forum).

I should practically import the inventor file (neutral format) into ansys, fix it and do the analysis. It seems convenient to me since we talk about the same figure that an external company asked us to realize the fem of a frame... .
I repeat that in the company we do not have 3d software because the designer is an external person.
Do any of you think I should address another software manufacturer? I never used one, I know that ansys is one of the best in the field for this I made contact with them. the guy, of course, told me that there's no need to trust analysis software applied by those who first produced the 3d, so the modules...

 

[Matteo]

First of all, thank you for your concern, and apologize for the section but I am not yet practical.
I have met ansys, practically as the analysis I have to carry out is in the linear and static field, it offers us:
- the basic package ansys designspace
-the module to create 3d or change them/adjust them as not always the designer (person outside the company) creates a good design to pass immediately to the analysis
- will follow me step by step in a project that I have to accomplish so as to create a personalized course

all to a figure much less than 20000 euros (I am not accurate because I do not know if it is correct to talk about prices on the forum).

I should practically import the inventor file (neutral format) into ansys, fix it and do the analysis. It seems convenient to me since we talk about the same figure that an external company asked us to realize the fem of a frame... .
I repeat that in the company we do not have 3d software because the designer is an external person.
Do any of you think I should address another software manufacturer? I never used one, I know that ansys is one of the best in the field for this I made contact with them. The guy, obviously, He told me that there is no need to trust analysis software applied by those who first produced the 3d, so to modules cad...

Mah... If I have understood your needs, I would say it is the appropriate package. I would like to know more about point 2: I hope they refer to the spaceclaim kernel cad and not to the few cad functions included in the "pre" ansys processor.

If the "made-to-measure" course is done well and provides total support, I would say that the cost can be:finger:

I don't understand the phrase I put in bold

 

[f_lambiase]

then, I gave him a .step file of a 3d frame that the designer realizes. they said that it was wrong because of the lines were not closed and therefore before the analysis it is necessary to have a package that allows to change the object made by the designer. Now I don't remember the name well but it should be something like ansys model, and he told me that wishing I can still create a 3d with such package (which we need little because we pay a designer, but as the designer, it is good, but it doesn't have engineering skills so it doesn't design a ready model and away for analysis, percui, such module I need to correct the drawings ).

I didn't understand what he said, but I think I realized that:
software created for the design3d to which only subsequently modules for fem analysis were added are not reliable because often such modules are not constantly updated are less immediate in use and slower. (of course)

 

[Matteo]

I don't remember the name well but it should be something like ansys model)

http://www.ansys.com/products/other+products/ansys+spaceclaim+direct+modelerif it is this go quiet, great purchase for an analyst fem, alone it is worth a few thousand €. basically it is an explicit cad (direct) superior to fusion of autodesk and many others :wink:
the important thing is that the course also catches this module, otherwise it is worth to deepen (look if we do spaceclaim courses).

software created for the design3d to which only subsequently modules for fem analysis were added are not reliable because often such modules are not constantly updated are less immediate in use and slower. (of course)

Let me say: This is a nonsense (besides incorrect). I don't find a word that can be shared in everything.. the same ansys that did not have a decent cad kernel is choir sheltered with the bundle mentioned above.. .
Of course these considerations are not addressed to you... ambasciator is not worth

 

[f_lambiase]

I remembered the name: ansys designmodeler I don't know if it's the same as the link you posted to me....I hope it's the same good :i

 

[cacciatorino]

However the reason why sometimes we do not recommend the integrated fems in the cad is very "particular":

having a tool like a fem at the hands of a designer, and easy to use, could lead the designer himself to the false conviction of being able to do analysis. Actually, doing fem analysis is very difficult and it's relatively easy to get wrong results without realizing it (for example, mesh with malformed elelemts in some areas to get the results high), if you don't have the right experience.

generally those who do the analyst do not plan, so he doesn't need a cad, and vice versa those who plan cannot do the analyst, so he doesn't need a fem. find the professional figure that can do both things at the level of competence required, it's very rare.

It is only for this reason that we do not recommend the integrated fems in the cad, which have only advantages, as it is obvious that it is (always that it is the full versions, and not the very limited ones that are inexisted sometimes in the basic package.

 

[Er Presidente]

However the reason why sometimes we do not recommend the integrated fems in the cad is very "particular":

having a tool like a fem at the hands of a designer, and easy to use, could lead the designer himself to the false conviction of being able to do analysis. Actually, doing fem analysis is very difficult and it's relatively easy to get wrong results without realizing it (for example, mesh with malformed elelemts in some areas to get the results high), if you don't have the right experience.

generally those who do the analyst do not plan, so he doesn't need a cad, and vice versa those who plan cannot do the analyst, so he doesn't need a fem. find the professional figure that can do both things at the level of competence required, it's very rare.

It is only for this reason that we do not recommend the integrated fems in the cad, which have only advantages, as it is obvious that it is (always that it is the full versions, and not the very limited ones that are inexisted sometimes in the basic package.

That's why I'm rare!!!! I'm not a designer and not an analyst!
:biggrin:

p.s.: You're right about the whole front! to each his profession, I "less"!

 

[cacciatorino]

That's why I'm rare!!!! I'm not a designer and not an analyst!
:biggrin:

p.s.: You're right about the whole front! to each his profession, I "less"!

You never know if you're serious or if you're serious.

Well, I'm alone with my new precision m4500, but I'm installing modeling (I'm making him the burden of being the first cad on this machine... :biggrin: ).

 

[Parametric_Ozzy]

That's why I'm rare!!!! I'm not a designer and not an analyst!
:biggrin:

p.s.: You're right about the whole front! to each his profession, I "less"!

When you do that, you almost scare me.

ps: the integrated sw are recommended when everyone does his job according to me...
It is clear that even to use a fem you have to have some training, as well as for cads, in my team I have practically only mechanical engineers, I am the only designer puppy... in fact I even if I so use it not use it.
for the series...
It's a dirty job, but someone has to do it.

currency also simulated creo (ex pro/mechanica), as Matteo says sometimes too often underestimated.

greetings

 

[Parametric_Ozzy]

You never know if you're serious or if you're serious.

Well, I'm alone with my new precision m4500, but I'm installing modeling (I'm making him the burden of being the first cad on this machine... :biggrin: ).

hunting eye because the ptc phallic logo hides a trojan.

 

[re_solidworks]

just to forget what they said about the "integrated" fems... in swx there is simulation that would be the old cosmos, while for the analysis of kinetodynamics there is motion that is an integrated reduction of adams, so that when committing errors come out on behalf of adams.

 

[Matteo]

I remembered the name: ansys designmodeler I don't know if it's the same as the link you posted to me....I hope it's the same good :i

is not the same product. Let us understand: it is not that dm is "more poor" of the module cad. are produced with different features: I am more likely to think that all the work on geoemtria is better to do it in an ambinete cad100% more than in an environment halfway between cad and pre-processor fem.

if it can be useful (with obvious confidentiality) see the model and understand what kind of simulations you need are available for a chatter (I promise I don't sell you any computing service).

without anything to remove to ansys, I suspect that with less expensive solutions (e.g. the fem module of inventor, which is nothing but the good algor) or in line with the proposed ansys package, but simpler to use (e.g. creosimulate, that I quote because now exists in version "stand alone", but wishing there would be others also the other integrated already mentioned).
Certainly ansys has the advantage of greater expandability towards "multiphysical" worlds, but it would be to understand what perspective of "expansion" has your ut.

 

[Matteo]

hunting eye because the ptc phallic logo hides a trojan.

Now that you have noticed I can no longer look at the ptc logo with the same eyes as before.. .

 

[The_Matrix]

if the integrated fem in the cad is poor then obvious it is not advisable. . .
but if it is a pottimo fem, (as for example nx adv sim) having it integrated is a cool... but...
but you can switch from 3d to fem and have the modified mesh (if possible) effortlessly... impatient. . .